Mudra Band’s Gesture Algorithm Focusing on Finger Movement Exceled in 1,000-User Test, Ready for Widespread Mobile OS Deployment
YOKNEAM ILLIT, ISRAEL, Nov. 14, 2023 (GLOBE NEWSWIRE) — Wearable Devices Ltd. (the “Company” or “Wearable Devices”) (NASDAQ:WLDS, WLDSW))))), a pioneering tech firm specializing in AI-powered touchless sensing wearables, today announced that its gesture recognition algorithm has passed accuracy performance testing on a carefully curated custom dataset for finger movement recognition based on first-time usage. The AI-based system is designed to produce models that address the three essential user experience requirements of accuracy, false positive recognition, and responsiveness.
The algorithm is universally compatible with the leading mobile operating systems. This includes support for TensorFlow Lite (tf-lite), a deep-learning framework for on-device inference, and Core ML, which facilitates the seamless integration of machine learning models into Apple devices. Versions based on TensorFlow Lite Micro are also available for embedded applications.
Leeor Langer, Chief Technology Officer and Co-Founder of Wearable Devices, commented, “We are pleased to surpass the 1,000-user test benchmark for our innovative gesture recognition algorithm, and are eagerly setting our sights on the next milestone of 10,000 users. Traditional input methods like the mouse and keyboard, while accurate and fast, lack mobility and hands-free operation, making them ill-suited for extended reality (XR) applications. Optical-based gesture recognition offers some mobility but has its limitations. Our AI-based algorithm circumvents these challenges, eliminating the need for cumbersome devices and field-of-view constraints. Combined with inertial measurement unit (IMU)-based air mouse solutions and biopotentials, we believe that we are building a superior system that optimizes the user experience for our products.”
